한국군사과학기술학회지 (Journal of the Korea Institute of Military Science and Technology)

  • 제24권3호
  • /
  • Pages.281-292
  • /
  • 2021
  • /
  • 1598-9127(pISSN)
  • /
  • 2636-0640(eISSN)
한국군사과학기술학회 (The Korea Institute of Military Science and Technology)
권호 표지
DOI QR코드

DOI QR Code

불균형데이터의 비용민감학습을 통한 국방분야 이미지 분류 성능 향상에 관한 연구

A Study on the Improvement of Image Classification Performance in the Defense Field through Cost-Sensitive Learning of Imbalanced Data

  • Jeong, Miae (Yulgok Yii, Republic of Korea Ship) ;
  • Ma, Jungmok (Department of Defense Science, National Defense University)
  • 투고 : 2020.09.17
  • 심사 : 2021.05.07
  • 발행 : 2021.06.05
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

With the development of deep learning technology, researchers and technicians keep attempting to apply deep learning in various industrial and academic fields, including the defense. Most of these attempts assume that the data are balanced. In reality, since lots of the data are imbalanced, the classifier is not properly built and the model's performance can be low. Therefore, this study proposes cost-sensitive learning as a solution to the imbalance data problem of image classification in the defense field. In the proposed model, cost-sensitive learning is a method of giving a high weight on the cost function of a minority class. The results of cost-sensitive based model shows the test F1-score is higher when cost-sensitive learning is applied than general learning's through 160 experiments using submarine/non-submarine dataset and warship/non-warship dataset. Furthermore, statistical tests are conducted and the results are shown significantly.

키워드

참고문헌

  1. S. W. Park, H. S. Park, Y. B. Kim, D. G. Kim, "Suspected Object Detection and Tracking methods for National Defence Surveillance," Communications of the Korean Institute of Information Scientists and Engineers, Vol. 36, No. 8, pp. 33-40, 2018.
  2. D. W. Choi, J. M. Ma, "Improvement of Ship Classification Based on Artificial Neural Network Considering Night and Low Visibility Marine Environment," Korean Journal of Computational Design and Engineer-ing, Vol. 24, No. 3, pp. 320-328, 2019. https://doi.org/10.7315/CDE.2019.320
  3. J. W. Lee, D. W. Nam, W. Y. Yoo, "A Study on Composition of Warship Type/Class Identification System," The Korea Institute of Information and Communication Engineering, Vol. 23, No. 2, pp. 607-610, 2019.
  4. M. A. Jeong, J. M. Ma, "Improvement of Submarine Detection Based on Mast Images Using An Ensemble Model of Convolutional Neural Networks," Journal of the Korea Institute of Military Science and Technology, Vol. 23, No. 2, pp. 115-124, 2020. https://doi.org/10.9766/KIMST.2020.23.2.115
  5. Log0, "Class Imbalance Problem," August 30, 2013, http://www.chioka.in/class-imbalance-problem/ (accessed July 10, 2020.).
  6. H. He, E. A. Garcia, "Learning from Imbalanced Data," IEEE Transactions On Knowledge And Data Engineering, Vol. 21, No. 9, pp. 1263-1284, 2009. https://doi.org/10.1109/TKDE.2008.239
  7. S. B. Kim, "Sampling Technique for Analysis of Imbalanced Data," February 20, 2020, https://www.youtube.com/watch?v=Vhwz228Vrlk (accessed June 28, 2020.).
  8. H. S. Jeong, C. W. Kang, K. K. Kim, "The Effect of Oversampling Method for Imbalanced Data," Journal of The Korean Data Analysis Society, Vol. 10, No. 4, pp. 2089-2098, 2008.
  9. K. N. Lee, J. T. Lim, K. S. Bok, J. S. Yoo, "Handling Method of Imbalance Data for Machine Learning : Focused on Sampling," The Korea Contents Society, Vol. 19, No. 11. pp. 567-577, 2019.
  10. H. C. Kim, "Adversarial Cost-Sensitive Learning for Class Imbalance Problem," M. E. Thesis, University of Gangwon, 2020.
  11. T. Fawcett, "Learning from Imbalanced Classes," August 25, 2016, https://www.svds.com/learning-imbalanced-classes/ (accessed July 10, 2020.).
  12. J. O. Kim, E. S. Hwang, J. M. Yoo, "Analysis of NCO Promotion Factors Using Imbalanced Data Processing," 12th International Army Modeling & Simulation Education Conference, p. 90, 2019.
  13. Solaris, "Deep Learning to Learn by Tensorflow," Youngjin.com, Seoul, 2018.
  14. K. Simonyan, A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," International Conference on Learning Representations, pp. 1-14, 2015.
  15. G. Saito, "Deep Learning from Scratch," Hanbit Media, Seoul, 2017.
  16. J. Brownlee, "How to Develop a Cost-Sensitive Neural Network for Imbalanced Classification," February 2, 2020, https://machinelearningmastery.com/cost-sensitive-neural-network-for-imbalanced-classification/ (accessed May 2, 2021.).
  17. S. J. Yoo et al., "Analysis of Big-Data Using Python," The Twenty-First Century, Paju, 2018.
  18. M. K. Seo, "Practical Data Processing and Analysis Using R," Gilbut Publishing, Seoul, 2014.